In the production of a printed circuit board (PCB), in a first (multi-step) stage a xe2x80x9cbare boardxe2x80x9d is prepared and in a second (multi-step) stage, various components are mounted on the board. The present invention relates to the final steps in the manufacture of the bare board, in which the bare board is coated with a protective layer prior to passing to the second production stage.
There are currently two types of components for attachment to the bare boards in the second stage referred to above: legged components e.g. resistors, transistors, etc., and, more recently, surface mount devices. Legged components are attached to the board by passing each of the legs through a hole in the board and subsequently ensuring that the hole around the leg is filled with solder. Surface mount devices are attached to the surface of the board by soldering with a flat contact area or by adhesion using an adhesive.
In the first stage referred to above, a board comprising an insulating layer, a conducting circuit pattern and conductive pads and/or through-holes is produced. The board may be a multi-layer board having more than one conducting circuit pattern positioned between insulating layers or may comprise one insulating layer and one conducting circuit pattern.
The through-holes may be plated through so that they are electrically conducting and the pads which form the areas to which the surface mount components will be attached in the subsequent component-attachment stage, are also electrically conducting.
The conductive areas of the circuit pattern, pads and through-holes may be formed from any conductive material or mixtures of different conductive materials. They are generally however, formed from copper. Since over time copper tends to oxidise to form a copper oxide layer with poor solderability, prior to passing to the second, component-attachment stage, a protective layer is coated over the pads and/or through-hole areas where it is desired to retain solderability to prevent formation of a poorly solderable surface layer of copper oxide.
While there is more than one way for preparing the bare boards, one of the most widely used processes for making the bare boards is known as the xe2x80x9csolder mask over bare copperxe2x80x9d (SMOBC) technique. Such a board generally comprises an epoxy-bonded fiberglass layer clad on one or both sides with conductive material. Generally, the board will be a multi-layer board having alternate conductive layers which comprise circuit pattern, and insulating layers. The conductive material is generally metal foil and most usually copper foil. In the SMOBC technique, such a board is obtained and holes are drilled into the board material using a template or automated drilling machine. The holes are then xe2x80x9cplated throughxe2x80x9d using an electroless copper plating process which deposits a copper layer on the entirety of the board: both on the upper foil surfaces and on the through-hole surfaces.
The board material is then coated with a light sensitive film (photo-resist), exposed to light in preselected areas and chemically developed to remove the unexposed areas revealing the conductive areas which are the plated through-holes and pads. Generally, in the next step, the thickness of the metal foil in the exposed areas is built up by a further copper electroplating step. A protective layer of an etch resist, which is usually a tin lead alloy electroplate composition is applied over the exposed and thickened copper areas.
The photo-resist is then removed exposing the copper for removal and the exposed copper surface is etched away using a copper etching composition to leave the copper in the circuit pattern finally required. In the next step, the tin-lead alloy resist is stripped away.
Since components will not be attached to the copper circuit traces, it is generally only necessary to coat the solder for attaching the components over the through-hole and pad areas but not the traces. Solder mask is therefore applied to the board to protect the areas where the solder coating is not required, for example using a screen printing process or photo-imaging technique followed by development and, optionally curing. The exposed copper at the holes and pads is then cleaned and prepared for solder coating and the protective solder coating subsequently applied, for example by immersion in a solder bath, followed by hot air leveling (HAL) to form a protective solder coating on the areas of copper not coated with solder mask. The solder does not wet the solder mask so that no coating is formed on top of the solder mask protected areas.
At this stage, the board comprises at least one insulating layer and at least one conductive layer. The conductive layer or layers comprise a circuit trace. The board also comprises a pad or pads and/or through-hole(s) protected from tarnishing by a layer of solder. A single conductive layer may comprise either a circuit trace or pad(s), or both. Any pads will be part of a conductive layer which is an outerlayer of a multi-layer board. The circuit traces on the board are coated with solder mask.
Such a board is ready to proceed to the second stage for attachment of the components. In this second stage, generally attachment of the components is achieved using solder: firstly a layer of solder paste (comprising solder and flux) is applied onto the boards, generally by printing and the components are positioned on the printed boards. The board is then heated in an oven to produce fusion of the solder in the solder paste, which forms a contact between the components and the board. This process is known as reflow soldering. Alternatively a wave soldering process is used in which the board is passed over a bath of molten solder. In either case additional solder is used over and above any protective solder coating.
The additional complications of attaching both legged components and the surface mount devices and the special requirements for mounting many small closely spaced components have resulted in increased demands on the surface protection coating for the conductive metal to which the components will be attached on the PCB""s. It is essential that the finish applied by the bare board manufacturer does not leave a pad with an uneven surface as this increases the risk of electrical failure. It is also essential that the protective coating does not interfere with the subsequent solder step, thereby preventing formation of a good, conducting bond between the bare board and components. An extra step in which the protective coating is removed would be undesirable.
As explained above, the conductive metal surfaces are generally formed of copper and the protective surface must be applied at the end of the first stage to prevent the formation of non-solderable copper oxide on the copper surfaces prior to the component attachment. This is particularly important because, generally speaking, the first stage and the second, component-attachment stage will be carried out at completely different sites. There may therefore be a considerable time delay between formation of conducting pads and/or through-holes and the component-attachment stage, during which time oxidation may occur. Therefore, a protective coating is required which will retain the solderability of conducting material and enable a soldered joint to be made when the components are attached to the bare boards.
The most common protection coating presently used is tin/lead solder, generally applied using the xe2x80x9cHALxe2x80x9d (hot air leveling) process, an example of which is described in detail above.
HAL processes are limited because it is difficult to apply the solder evenly and the thickness distribution produced by the use of HAL processes makes it difficult to reliably attach the very small and closely spaced components now being used.
Several replacement treatments for the HAL coating of a solder layer are being introduced. The coatings must enable formation of a reliable electrical contact with the component. They should also be able to stand up to multiple soldering steps. For example, as described above, there are now both legged and surface mount components for attachment to the bare boards and these will generally be attached in at least two soldering operations. Therefore, the protective coatings must also be able to withstand at least two soldering operations, so that the areas to be soldered in a second operation remain protected during the first operation.
Alternatives to the tin/lead alloy solder used in the HAL process, which have been proposed include organic protection, immersion tin or tin/lead plating and nickel/gold plating. In the nickel/gold process electroless plating of the copper surfaces is carried out in which a primer layer of nickel is applied onto the copper followed by a layer of gold. This process is inconvenient because there are many process steps and in addition, the use of gold results in an expensive process.
Organic protection for the copper pads during storage and assembly prior to soldering have also been effected using flux lacquer. Its use is generally confined to single-sided boards (i.e. boards which have conductive pads on only one side). The coating is generally applied by dip, spray or roller coating. Unfortunately, it is difficult to provide a consistent coating to the board surfaces so limited life expectancy, due to the porosity of the coating and to its inconsistent coating thickness, results. Flux lacquers have also been found to have a relatively short shelf life. A further problem is that in the component-attachment stage, if reflow soldering is to be used to attach the components, the components are held in place on the underside of the boards with adhesive. In cases where the flux lacquer is thick, the adhesive does not bond the component directly to the printed board, but instead forms a bond between the adhesive and the lacquer coating. The strength of this bond can drop during the fluxing and soldering step causing components to be lost during contact with the solder baths.
One other alternative currently being used is passivation/protection treatment based upon the use of imidazoles or triazoles in which copper-complex compounds are formed on the copper surface. Thus, these coatings chemically bond to the surface and prevent the reaction between copper and oxygen. However this process is disadvantageous because it tends to be inadequate for withstanding successive soldering steps so that the high temperatures reached in a first soldering step tend to destroy the layer which cannot withstand a subsequent soldering operation needed to mount further components. One example of such a process is given in EP-A-0428383, where a process is described for the surface treatment of copper or copper alloys comprising immersing the surface of copper or copper alloy in an aqueous solution containing a benzimidazole compound having an alkyl group of at least 3 carbon atoms at the 2-position, and an organic acid.
Processes are also known which provide coatings using compositions which comprise silver.
The three common complexing systems for electroless silver plating processes are either ammonia-based, thiosulphate-based or cyanide-based.
The ammonia systems are disadvantageous because the ammonia-containing silver solutions are unstable and explosive azides may tend to form. Thiosulphate systems are disadvantageous for use in the electronics industry because sulphur compounds in the silver coatings formed result in poor solderability so that in the subsequent component-attachment step, a poor electrical contact may be formed between the bare board and the component.
The cyanide-based systems are disadvantageous due to the toxicity of the plating solutions.
In U.S. Pat. No. 5,318,621 an electroless plating solution containing amino acids as rate enhancers for depositing silver or gold onto a nickel coating overlying copper on a circuit board is disclosed. It is described that neither gold nor silver electroless plating baths based on thiosulphate/sulphate will plate directly onto copper because the copper rapidly dissolves without allowing a silver or gold coating to form. In the introduction of the reference, xe2x80x9cMetal Finishing Guidebook and Directoryxe2x80x9d (1993 edition), silver plating solutions comprising silver nitrate, ammonia and a reducing agent such as formaldehyde are mentioned.
U.S. Pat. No. 4,863,766 also discloses electroless silver plating, using a cyanide-based plating solution. In Metal Finishing (1983) 81(i), pp 27-30 Russev described immersion silvering of copper powder from a plating solution containing silver nitrate and a nitrogen complexing agent. In Metal Finishing (1960) August, p 53 Geld described a silver coating process involving an initial bright dip of the brass or copper substrate, followed by a silver plating step in which a thick coating of silver is plated from a solution of silver nitrate and potassium iodide. The process is for plating of electrical contacts to increase conductivity.
In JP-A-04-110474 a base material is plated with silver, dried and subsequently treated with a mercaptan compound to prevent tarnish.
In DE-C-4316679 base metals such as copper are coated with palladium in a two-step process including a first step in which the surface is contacted with a bath containing a palladium salt and an oxidizing agent, and in the second step with a bath containing a palladium salt, a complexing agent and formic acid or formic acid derivative. The latter bath may also contain stabilizers for the bath itself, which stabilize the bath against decomposition or xe2x80x9cplating-outxe2x80x9d. It is suggested that the copper substrate should previously be etched using a non-bright etch bath including persulphate. However, such pretreatment steps tend to produce relatively porous coatings. The inventors there minimize the porosity of the coating by using the two-step process in the first of which a very thin coating is formed. This reference warns against using silver as corrosion protection due to migration.
The present invention relates to a displacement immersion metal plating in which a more electropositive metal displaces a less electropositive metal at the surface to be coated. Ions of the more electropositive metal oxidize the substrate metal. A displacement plating process differs from an electroless process because the silver coating forms on the surface of a metal by a simple displacement reaction due to the relative electrode potentials of the oxidisable metal of the surface to be protected and of the silver ions respectively.
It is reported in for example xe2x80x9cModern Electroplatingxe2x80x9d by F. A. Lowenheim, published by J. Wiley and Sons (1963) that silver will plate by displacement on most base metals but that immersion plated silver is poorly adherent. F. A. Lowenheim there suggests that when electroplating base metals with silver, it is necessary to deposit first a thin film of silver on the work piece from a high-cyanide strike bath to ensure good adhesion of the subsequent electroplated silver layer.
The present invention aims to provide an alternative to the solder protection coating for the copper or other conducting surfaces of bare boards which require protection from tarnishing between bare board manufacture and the component-attachment stage.
In accordance with the present invention, there is provided a method for coating a PCB comprising an insulating layer and a conducting layer, with metal pads and/or through-holes in which the pads and/or through-holes are provided with an anti-tarnish coating, the process comprising contacting the pads and/or through-holes with a bright-etch composition in a bright-etch step; and subsequently immersion plating the etched pads and/or through-holes in a metal-plating step to form solderable plated metal surfaces.